Sensitization to subtropical grass pollens in patients with seasonal allergic rhinitis from Bahia Blanca, Argentina.

BACKGROUND: Grass pollinosis is an important contributor to allergic diseases, with varying patterns and frequency of allergens according to the geographical location studied. Our study aims to provide a better understanding of subtropical grass pollinosis in Argentinian patients with seasonal allergic rhinitis.
METHODS: We conducted a retrospective cross-sectional study involving 894 patients with seasonal allergic rhinitis from Bahía Blanca, Argentina. Skin prick tests were performed to selected pollen species belonging to three subfamilies of the Poaceae genera. Frequency of sensitization to specific grass pollen extracts, cross-reactivity of allergens assessed by skin prick test, and possible associations between allergen extracts and asthma or allergic conjunctivitis were analyzed.
RESULTS: Sensitization to the Pooideae subfamily was the most frequent, encompassing 86.8% (CI: 84.4%-88.9%) of the studied population. Positive reactions to allergen extracts from the Chloridoideae and the Panicoideae subfamilies showed smaller papule size than allergen extracts from the Pooideae subfamily (χ2(5) = 83.75, p < 0.001). Patients with a positive skin prick test (SPT) to a specific extract were more likely to present some degree of cross-reactivity to the remaining pollens when compared to patients with negative SPT to the same specific extract. Even though the proportion of patients presenting with asthma (46.9%) was higher than those with conjunctivitis (22.6%), there was only a statistically significant association between sensitization to Festuca arundinacea (φ = 0.089, p = .009), Phalaris arundinacea (φ = 0.074, p = .032) and Paspalum notatum (φ = 0.070, p = .038) and the presence of conjunctivitis.
CONCLUSIONS: Our results suggest a high frequency of sensitization to grass pollen extracts from the Poaceae family among patients with seasonal allergic rhinitis. Overall, sensitization to the Pooidae subfamily was the most common, where Phalaris arundinacea presented the highest frequency.

Background

Grass pollinosis is an important contributor to allergic diseases, with varying patterns and frequency of allergens according to the geographical location studied. In general, there is an inverse distribution of temperate and subtropical grasses, with species of the latter being more abundant the closer to the equator. More than 95% of allergy-relevant grass species belong to three subfamilies of the Poaceae family: Pooideae, Chloridoideae, and Panicoideae.Pooideae grasses are most frequent on temperate climate zones, while Chloridoideae can be found on temperate and sub-tropical grasslands, and Panicoideae is mainly found on subtropical grasslands.

Identification of sensitization patterns and airborne allergen counts are clinically relevant to provide a more accurate diagnosis in allergic conditions such as seasonal allergic rhinitis and have the potential for treatment through immunotherapy. However, climate change as a result of global warming is altering the distribution, timing, and allergenicity of grass pollens, with an expected increase of subtropical grasses. Furthermore, with the widening of subtropical climate zones, and the increasing population size living in these regions, it is expected that allergy to subtropical grasses will play a key role in the exacerbation of allergic diseases in the future.5, 6

Much of what is known about the allergenic components of grasses, distribution, and patterns of sensitization primarily comes from developed regions, where temperate grass pollinosis is more prevalent. Data from Latin America in this regard are relatively sparse. In Argentina, there is a characteristic geographic distribution and high-frequency airborne count of pollens. However, to the best of our knowledge, there are no studies that characterize the frequency of sensitization to specific extracts. Our study aims to provide a better understanding of subtropical grass pollinosis in Argentinian patients with seasonal allergic rhinitis, by identifying the frequency of sensitization to specific grass pollen extracts, cross-reactivity of allergens assessed by skin prick test, and the possible associations between allergen extracts and asthma or allergic conjunctivitis.

Methods

Study design and procedure

We conducted a retrospective cross-sectional study involving 894 patients with seasonal allergic rhinitis from Bahía Blanca, Argentina (Table 1). Participation was restricted to patients presenting symptoms for at least two consecutive years, with at least one positive skin prick test (SPT) to any of the six allergen extracts of the Poaceae genera studied and had resided at the city of Bahia Blanca for at least five years. Skin prick tests (SPTs) were performed at the Instituto de Alergia e Inmunología del Sur (Bahía Blanca, Buenos Aires, Argentina) from November 2010 to November 2016 using GREER® GreerPick™ Skin Testing Devices. Allergen extracts were provided by Alergo Pharma SRL and produced by Greer Labs (GREER®).

Table 1

Demographic and clinical information of surveyed population (n = 894)

Characteristics	Value n (%)
Age (mean)	32.2 (17.0)
Gender
Male	457 (51.1)
Female	437 (48.9)
Asthma	419 (46.9)
Allergic conjunctivitis	202 (22.6)

Skin prick tests were considered positive when the papule diameter was equal to or greater than 3 mm from the negative control papule. The selected pollen species belonged to three subfamilies of the Poaceae genera: the Pooideae subfamily (Lolium perenne, Festuca arundinacea and Phalaris arundinacea), the Panicoideae subfamily (Paspalum notatum and Sorghum halepense) and the Chloridoideae subfamily (Cynodon dactylon) (Table 2). Species were selected due to their representativeness of the area of study, except for Paspalum notatum, which was selected due to its taxonomic affinity to Paspalum quadrifarium.9, 10 Even though Paspalum quadrifarium is the regional representative specie, there was no extract available for testing.

Table 2

Species used in this study and related species native from the region of Bahía Blanca.

Species	Subfamily	Statusa	Native relatives
Lolium perenne	Pooideae	Exotic	N/A
Festuca arundinaceab	Pooideae	Exotic	F. pampeana
			F. ventanicola
Phalaris arundinacea	Pooideae	Exotic	P. platensis
Cynodon dactylon	Chloridoideae	Exotic	N/A
Paspalum notatum	Panicoideae	Native	P. quadrifariumc
			P. dilatatum
Sorghum halepense	Panicoideae	Exotic	N/A

Notes: Species names correspond to the updated nomenclature from the Flora del Conosur Database (IBODA). N/A, not applicable.

The term ‘Status’ refers to evolutive origin in botany.

Festuca arundinacea was used instead of the old name Festuca elatior.

Allergen extracts not available

Ethical considerations

This study was approved by the ethics committee: Comité de ética e Investigación en Seres Humanos (CEISH), in accordance to the principles established by the declaration of Helsinki.

Statistical analyses

Throughout data analyses, SPT was manipulated either as a continuous (papule's diameter in mm) or dichotomic (positive or negative) variable. Analyses treating SPT as a continuous variable were intended to evaluate the degree of reactivity upon a certain dependent variable, whilst tests analyzing SPT as a dichotomic variable were included to give a better clinical utility.

Descriptive statistics for proportions of patients with a positive SPT for the selected species were represented using intervals of 95% confidence. Cochran's Q test was used to determine the statistical significance of such proportions. A Cochran's Q test was also performed considering the SPT response to pollen extracts by subfamilies. Difference of papule's size between allergen extracts were analyzed using Friedman test.

Thereafter, a Kendall's tau b to assess the association between the degree of reactivity of each extract and the SPT response of the other species was performed. A Chi-square test for association was reproduced as well, for the same purpose. We employed the same tests to determine the association between SPT to the allergen extracts and the presence of asthma and allergic conjunctivitis.

All the data was analyzed using SPSS, version 24.0 software (SPSS Inc., Chicago, IL, USA). We performed Fisher's exact test were necessary. A p-value of less than 0.05 was considered statistically significant for all tests.

Results

Sensitization to the Pooideae subfamily was the most frequent, 86.8% (CI: 84.4%–88.9%) of the studied population, compared to the other subfamilies: the Panicoideae species disclosed a 69.6% (CI: 66.5%–72.5%) positive while the Chloridoideae subfamily presented a 48.1% (CI: 44.8%–54.1%) positive. The Cochran's Q test suggested that the prevalence in the three groups was different (χ2(3) = 319.11, p < .01). The same analysis, considering species independently of their subfamilies, was statistically significant as well (χ2(5) = 69.31, p < .001). Descriptive statistics of allergy prevalence by species from the three subfamilies are best depicted in Fig. 1 and Supplemental Appendix Table S1.

Fig. 1

Prevalence of sensitization to grass pollen allergens in patients with seasonal allergic rhinitis. Values shown are mean ± SE in a percentage scale

Mean difference and distribution

Positive reactions to allergens from the Chloridoideae and the Panicoideae subfamilies showed lower reactivity than allergens from the Pooideae subfamily. Differences in the sizes of papule after the allergen extract administration were statistically significant (χ2(5) = 83.75, p < .001). The distribution of the measures of the SPT of every allergen extract are depicted in Fig. 2. Notice that all allergen extracts but Festuca arundinacea and Phalaris arundinacea, showed a longer segment between the third quartile and the median than the segment within the median and first quartile. This description suggests a relatively uniform distribution for all species except for Festuca arundinacea and Phalaris arundinacea, which were left-skewed. Although the prevalence to the Pooideae subfamily of pollens seemed to be higher, the sensitivity of positive cases to the Panicoideae and the Chloridoideae subfamilies exhibited more outliers.

Fig. 2

Papule size distribution by pollen grass species. Median (midline), box (25th and 75th percentiles), whiskers (extrema), ° (outlier) and * (extreme value)

Correlations cross-reactivity

SPT for each of the allergen extract reported cross-reactivity between all species, as assessed by Kendall's τb. The most negligible correlation was appreciated between Cynodon dactylon and Phalaris arundinacea (τb = 0.160, p = p < .001), while the best positive correlation was measured between Phalaris arundinacea and Festuca arundinacea (τb = 0.330, p < .001), although still low. Moreover, Chi-squared tests revealed similar results, denoting that patients with positive SPT to any allergen extract were more likely to have cross-reactivity to all the other extracts when compared to patients with a negative SPT to the same specific allergen extract. Proportions of patients presenting cross-reactivity and the correlation strength assessed by Phi coefficient are best portrayed in Fig. 3.

Fig. 3

Cross-reactivity between allergen extracts of selected species. Five extracts of allergens are represented on each pentagon. By visualizing each of these five extracts independently, it is assumed that absolutely all patients have a positive SPT for each extract indicated at the vertex of the pentagon. Similarly, visualizing each of these five extracts independently, the proportions of patients with positive and negative SPT to the pollen noted below the pentagon can be observed. All the proportions presented in the graph revealed a statistically significant association (p < .05) through a Chi-square test, and the strength of such association was measured by the Phi coefficient indicated in each vertex of the pentagon according to the allergen extract. Lol P, Lolium perenne; Fes A, Festuca arundinacea; Pha A, Phalaris arundinacea; Cyn D, Cynodon dactylon; Pas N, Paspalum notatum; Sor H, Sorghum halepense; SPT, skin Prick test; Φ, Phi coefficient

Overall, the proportion of patients also presenting with asthma was higher in the negative SPT group, except for Paspalum notatum. On the other hand, the proportion of patients with conjunctivitis was higher in the positive SPT group (Fig. 4). However, Kendall's τb revealed no correlation between the degree of reactivity to pollen extracts and asthma, whilst correlations for allergic conjunctivitis were significant for Festuca arundinacea (τb = 0.088, p = .003) and Phalaris arundinacea (τb = 0.066, p = .025). Analyzing the SPT as a nominal variable, Chi-squared tests for association did not reveal a statistically significant correlation for asthma. With respect to conjunctivitis, there was a positive relationship for Festuca arundinacea (φ = 0.089, p = .009), Phalaris arundinacea (φ = 0.074, p = .032) and Paspalum notatum (φ = 0.070, p = .038), though, Phi coefficient identified it as negligible.

Fig. 4

Proportions of patients with asthma symptoms and allergic conjunctivitis by SPT response to allergen extracts of selected species. Proportions are expressed as percentages. Lol P, Lolium perenne; Fes A, Festuca arundinacea; Pha A, Phalaris arundinacea; Cyn D, Cynodon dactylon; Pas N, Paspalum notatum; Sor H, Sorghum halepense

Discussion

Data on sensitization to grass pollen extracts primarily comes from developed regions such as North America and Europe, with overall sensitization rates around 15–20% among the general population. However, these rates tend to vary depending on the geographical location and specific sample analyzed. In our study, we found patients with seasonal allergic rhinitis to be most frequently sensitized to extracts belonging to the Pooideae subfamily (86.8%), followed by Panicoideae (69%), and Chloridoideae (48.1%). Although data from Latin America in this regard is relatively sparse, our results suggest that there is a higher frequency of sensitization to grass allergens when compared to the rate of sensitization found on the general population from other countries of the region (20–30%).11, 12, 13, 14

When assessing specific extracts from each subfamily, sensitivity to Phalaris arundinacea was the most common (62.2%), followed closely by the other extracts of the Pooideae subfamily (Festuca arundinacea, and Lolium perenne). We also found lower reactivity to allergens from the Chloridoideae (Cynodon dactylon) and Panicoideae (Paspalum notatum, Sorghum halepense) subfamilies when compared to the Pooideae subfamily, with a statistically significant difference in the size of papules size (χ2(5) = 83.75, p < .001). Such observation contrasts with findings from another subtropical region, were Cynodon dactylon, and Paspalum notatum exhibited greater sensitivity measured by SPT than Lolium perenne. In Europe, differences in level of reactivity have also been observed and appear to be related to the patient's country of origin. For instance, a lower degree of SPT response to Cynodon dactylon was found when compared to temperate grass pollens extracts, such as Lolium Perenne. These observations further exemplify the regional differences seen in allergic sensitivity to different types of subtropical grass pollens.

In Poaceae, although species belonging to different sub-families are characterized by distinct allergen subsets, there is a considerable degree of cross-reactivity between many species, even between temperate and subtropical grasses. For example, in IgE cross-inhibition assays performed in patients with grass pollinosis in India, five locally abundant subtropical grasses: Cynodon, Cenchrus, Imperata, Pennisetum, and Sorghum, revealed asymmetric IgE cross-inhibition consistent with the subfamily of grass pollens. Albeit not a direct comparison, we found that patients with a positive SPT to a specific extract were more likely to present some degree of cross-reactivity to the remaining pollens when compared to patients with negative SPT to the same specific extract. These findings might be the result of structurally-related proteins (e.g. profilins, polcalcins, lipid-transfer proteins) responsible for cross-reactivity among different allergenic sources.18, 19 However, since our analyses revealed that such associations were weak, it can be concluded that regardless of SPT response to a certain extract, patients still need to be screened for all the remaining pollens. Further studies through IgE cross-inhibitions assays should be performed to determine the exact extent of the observed cross-reactivity.

The clinical implications of grass pollen allergy are mainly related to the triggering of allergic diseases in sensitized patients. Peaks in airborne grass pollen in late spring and early summer can give rise to exacerbations of allergic rhinitis, allergic conjunctivitis, and asthma. For instance, a study in Australia found the levels of airborne grass pollen to be a strong independent non-linear predictor of hospital admissions related to asthma. In our study, even though roughly 40% of patients with seasonal allergic rhinitis also presented asthma, its proportion was slightly higher in patients with negative SPT in some grass species when evaluating each specific extract. However, the correlation between asthma and SPT response was not statistically significant. On the other hand, even though allergic conjunctivitis was less frequent than asthma in our sample, we found a statistically significant association between its presence and sensitization to Festuca arundinacea, Phalaris arundinacea and Paspalum notatum extracts, with frequencies of sensitization around 25%. In comparison, a study of the most common allergens in Northern Greece found that 58.4% of patients with allergic conjunctivitis presented a positive SPT to a grass mix.

Finally, there are several limitations in our study that require mentioning. First, our results were restricted to the availability of allergens representing the vegetation of our region. Second, since an extract to Paspalum quadrifarium was not available, which is a true representative of the region, Paspalum notatum was used instead, leading to indirect comparisons despite its taxonomic affinity. Moreover, cross-reactivity was assessed by SPT response, however, cross-inhibition IgE assays should be performed to formally establish cross-reactivity. Finally, since we did not screen for other allergen sensitizations besides pollen, airborne fungus and dust mites, we cannot guarantee that patients were monosensitized exclusively to pollen, and thus, we cannot explain if other antibodies take part in the cross-reactivity since antibodies to other allergens not included in this study may also cross-react to the allergens tested in the study if such antigens share similar epitopes. To the best of our knowledge, this is the first study to assess specific grass pollinosis among patients with seasonal allergic rhinitis from Bahía Blanca, Argentina.

Conclusion

Our results suggest a high frequency of sensitization to grass pollen extracts from the Poaceae family among patients with seasonal allergic rhinitis. Overall, sensitization to the Pooidae subfamily was the most common, where Phalaris arundinacea presented the highest frequency (62.2%). Positive reactions to allergens from the Chloridoideae and the Panicoideae subfamilies exhibited lower SPT response than allergens from the Pooideae subfamily. Even though the proportion of patients presenting with asthma (46.9%) was higher than those with conjunctivitis (22.6%), there was only a statistically significant association between sensitization to Festuca arundinacea, Phalaris arundinacea, and Paspalum notatum, with conjunctivitis.

Declarations

Ethics approval and consent to participate

This study was approved by the ethics committee Comité de ética e Investigación en Seres Humanos (CEISH) in accordance to the principles established by the declaration of Helsinki.

Consent for publication

Not applicable.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Funding

This study was partially supported by an unrestricted grant from Universidad Espíritu Santo. The sponsor had no role in the design of the study or in the collection, analysis, and interpretation of data.

Competing interest

The authors declare no relevant conflicts of interest related to this work.

Author's contributions

Authors have made substantial contributions to conception and design, acquisition, analysis and interpretation of data, have been involved in drafting the manuscript or revising it critically for important intellectual content, and given final approval of the version to be published. GDR, LBB, VV participated in the recollection of data. EV performed the statistical analyses. GDR, EV, MF, ICO participated in the interpretation of analyses, figures, tables, and drafting of the manuscript. EV, MF collaborated in revising and improving the final version of the manuscript. All authors read and approved the final version.